Electric power complaint work order multi-label text classification method based on self-attention mechanism

The invention provides an electric power complaint work order multi-label text classification method based on a self-attention mechanism. The dependence between labels is effectively coded by using the self-attention mechanism. The method may efficiently encode dependencies between tags using a self-attention mechanism. The method not only effectively solves the classification problem of the powercomplaint work order text, but also overcomes the defect that the dependency relationship between labels is ignored in the prior art. 本申请提出了基于自注意力机制的电力投诉工单多标签文本分类方法，使用自注意力机制对标签之间的依赖性进行有效编码。该方法可使用自注意力机制对标签之间的依赖性进行有效编码。不仅有效解决了电力投诉工单文本的分类问题，还克服了现有技术忽略标签之间依赖关系的缺点和不足。